Liquid fueled engines have proven quite safe since the Apollo pogo incident over 40 years ago, with only one in-flight shutdown of one engine in the Shuttle program, and that due to a faulty sensor. Solids are, i think it's fair to say, somewhat less safe. Which is more expensive, a new expendable engine or refurbishing a reusable one? For solids, refurbishing an SRB is extremely labor-intensive, and, when all costs are considered, probably significantly more expensive than building a new one. For example, every part must be abrasive-blasted to remove all coating and plating, then magnafluxed and replated before reassembly. Refurbishing the Shuttle Main Engines is more economical. Now, the RS-68 is an expendable liquid-fuel engine with remarkable performance and relatively low cost. But if a future shuttle engine is designed to go a substantial number of flights between overhauls, if the vehicle lands on a runway rather than in salt water, and if the engine is designed to simplify the overhaul procedure, then I think a reusable liquid-fueled engine can definitely be less expensive, in total life cycle cost, then an expendable one that must be replaced with each flight. To verify that this is true we need a reusable, unmanned liquid-fueled technology demonstrator, like the X-34, X-33, or DC-X.